Editor's Summary

9 July 2009

Cellular information processing

While naked DNA has a relatively static and easy to grasp information capacity — 2 bits per nucleotide—reversible chemical modification at multiple sites in even a single protein encodes a potentially large and so far untractable amount of information. Here Matthew Thomson and Jeremy Gunawardena reduce the 3 2n nonlinear differential equations describing dynamic phosphorylation at n sites on a given protein (n varying from less than 7 in bacteria to more than 150 in eukaryotes) to just two algebraic equations. The method allows them to estimate the information capacity of a signalling protein as a function of varying amounts of modifying enzymes (kinases and phosphatases). Algebraic geometry could extend the method to diverse and parallel enzymatic modifications such as those governing the 'histone code' of gene regulation.

Letter: Unlimited multistability in multisite phosphorylation systems

Matthew Thomson & Jeremy Gunawardena

doi:10.1038/nature08102

First paragraph | Full Text | PDF (397K) | Supplementary information